GAO Report On Reverse Mortgages July 2009

United States Government Accountability Office
GAO Report to Congressional Committees
July 2009
REVERSE
MORTGAGES
Policy Changes Have

Had Mostly Positive
Effects on Lenders
and Borrowers, but
These Changes and
Market Developments
Have Increased HUD’s
Risk
GAO-09-836
July 2009
REVERSE MORTGAGES
Accountability Integrity Reliability
Highlights
Highlights of GAO-09-836, a report to
Policy Changes Have Had Mostly Positive Effects on
Lenders and Borrowers, but These Changes and
Market Developments Have Increased HUD's Risk
congressional committees
Why GAO Did This Study What GAO Found

Reverse mortgages—a type of loan On the basis of a survey of HECM lenders, GAO estimates that taken together,
against home equity available to HERA’s changes to the HECM loan limit and origination fee calculation have
seniors—are growing in popularity. had a positive to neutral influence on most lenders’ plans to offer HECMs.
A large majority of reverse Other factors, such as economic and secondary market conditions, have had a
mortgages are insured by the mixed influence. Although economic conditions have had a positive influence
Department of Housing and Urban
on about half of lenders’ plans to offer HECMS, secondary market conditions
Development (HUD) under its
Home Equity Conversion Mortgage have negatively influenced about one-third of lenders. GAO also estimates that
(HECM) program. the HERA changes have had little to no influence on most lenders’ plans to
offer non-HECM reverse mortgages.
The Housing and Economic
Recovery Act of 2008 (HERA) HERA’s provisions will affect borrowers in varying ways depending on home
made several modifications to the value and other factors. The changes to HECM origination fees and loan limits
HECM program, including changes are likely to change the up-front costs and the loan funds available for most
in how origination fees are new borrowers. GAO’s analysis of data on HECM borrowers from 2007 shows
calculated and an increase in the that if the HERA changes had been in place at the time, most would have paid
loan limit. The Act directed GAO to less or the same amount in up-front costs, and most would have had more or
examine (1) how these changes the same amount of loan funds available. For example, about 46 percent of
have affected lenders’ plans to
offer reverse mortgages, (2) how
borrowers would have seen a decrease in up-front costs and an increase in
the changes will affect borrowers, available loan funds. However, 17 percent of borrowers would have seen an
and (3) actions HUD has taken to increase in up-front costs and a decrease in available loan funds.
evaluate the financial performance
of the HECM program. To address HUD has enhanced its analysis of HECM program costs, but less favorable
these objectives, GAO surveyed a house price trends and loan limit increases have increased HUD’s risk of
representative sample of HECM losses. HUD has updated its cash flow model for the program and plans to
lenders, analyzed loan-level HECM conduct annual actuarial reviews. Although the program historically has not
data, and reviewed HUD estimates required a subsidy, HUD has estimated that HECMs made in 2010 will require
and analysis of HECM program a subsidy of $798 million, largely due to more pessimistic assumptions about
costs. long-run home prices. In addition, the higher loan limit enacted by HERA may
increase the potential for losses. To calculate the amount of funds available
What GAO Recommends to a borrower, lenders start with a limiting factor of either the home value or,
GAO makes no recommendations in if the home value is greater than the HECM loan limit, with the loan limit. For
this report. HUD concurred with the loans that are limited by the home value, the loan amount and the home value
report’s findings. are closer together at the point of origination, which makes it more likely that
the loan balance could exceed the home value at the end of the loan. In
contrast, for loans that are limited by the HECM loan limit, there is initially a
greater difference between the home value and the loan amount, making it
less likely that the loan balance will exceed the home value at the end of the
loan. The increase in the HECM loan limit may increase HUD’s risk of losses
by reducing the proportion of loans that are limited by the HECM loan limit.
View GAO-09-836 or key components.

For more information, contact Mathew J.
Scirè at (202) 512-8678 or sciremj@gao.gov.

Contents
Letter 1
Background 3
Most HECM Lenders View the Overall Effect of the HERA
Provisions as Neutral or Positive for Their Reverse Mortgage
Business 11
HERA Provisions Will Affect Borrower Costs and Loan Amounts
Differently Depending on Home Value and Other Factors 20
HUD Has Enhanced Its Analysis of HECM Program Costs but
Changes in House Price Trends and Higher Loan Limits Have
Increased HUD’s Risk of Losses 27
Agency Comments and Our Evaluation 37
Appendix I Objectives, Scope, and Methodology 38
Appendix II Impact of Loan Limit Increase in the American

Recovery and Reinvestment Act of 2009 on HECM
Lenders 45
Appendix III Effect of the Housing and Economic Recovery Act of

2008 on Up-front Costs for HECM Borrowers 46
Appendix IV GAO Contact and Staff Acknowledgments 48
Tables
Table 1: Change in Up-front Costs for $300,000 Houses in Various
Locations 21
Table 2: Survey Population and Sample Dispositions 39
Table 3: Universe of 2007 HECMs by Home Value and Maximum
Claim Amount Category 43
Figures
Figure 1: Comparison of 30-Year Forward and Reverse Mortgages 4
Page i GAO-09-836 Reverse Mortgages

Figure 2: Number of HECMs Insured Annually, Fiscal Years 1990
through 2008 5
Figure 3: Potential Liability of Active HECM Loans, Fiscal Years
2006 through 2008 6
Figure 4: Number of HUD-Approved Lenders Originating HECMs,
Fiscal Years 2003 through 2008 7
Figure 5: FHA Process to Determine the Maximum Claim Amount
and the Amount of Loan Funds Available 9
Figure 6: Influence of HERA’s Provisions on Loan Limits and Fees
and Other Factors on Lenders’ Plans to Offer HECMs 13
Figure 7: Influence of HERA’s Provisions on Loan Limits and Fees
and Other Factors on Lenders’ Plans to Offer Non-HECM
Reverse Mortgages in 2009 19
Figure 8: Number of 2007 HECM Borrowers Affected by HERA
Provisions 24
Figure 9: Average Changes in Maximum Claim Amounts, Up-front
Costs, and Loan Funds Available for 2007 HECM
Borrowers 25
Figure 10: HECM Credit Subsidy Rates, Fiscal Years 2006 through
2010 31
Figure 11: Loan Liability Guarantee for the HECM Program, Fiscal
Years 2006 through 2008 32
Figure 12: Illustration of How Increasing the HECM Loan Limit
Could Increase HUD’s Losses 34
Figure 13: Percentage of HECMs with Maximum Claim Amounts
Limited by the Program Limit 36
Figure 14: Influence of ARRA’s Increase to Loan Limits on Lenders’
Plans to Offer Reverse Mortgages 45
Figure 15: Influence of ARRA’s Increase to Loan Limits on
Consumer Demand for HECMs 45
Figure 16: Changes to Up-front Costs for Borrowers Not Affected
by HERA’s Change in Loan Limit 47
Figure 17: Changes to Up-front Costs for Borrowers Affected by
HERA’s Change in Loan Limit 47
Page ii GAO-09-836 Reverse Mortgages

Abbreviations
ARRA American Recovery and Reinvestment Act of 2009

CBO Congressional Budget Office
FCRA Federal Credit Reform Act
FHA Federal Housing Administration
GI/SRI General Insurance and Special Risk Insurance
HECM Home Equity Conversion Mortgage
HERA Housing and Economic Recovery Act of 2008
HMBS HECM Mortgage Backed Security
HUD Department of Housing and Urban Development
LLG liability for loan guarantees
MBA Mortgage Bankers Association
MMI Mutual Mortgage Insurance
NRMLA National Reverse Mortgage Lenders Association
OIG Office of the Inspector General
This is a work of the U.S. government and is not subject to copyright protection in the
United States. The published product may be reproduced and distributed in its entirety
without further permission from GAO. However, because this work may contain
copyrighted images or other material, permission from the copyright holder may be
necessary if you wish to reproduce this material separately.
Page iii GAO-09-836 Reverse Mortgages

Washington, DC 20548
July 30, 2009
The Honorable Christopher Dodd

Chairman
The Honorable Richard Shelby
Ranking Member
Committee on Banking, Housing, and Urban Affairs
United States Senate
The Honorable Barney Frank

Chairman
The Honorable Spencer Bachus
Ranking Member
Committee on Financial Services
House of Representatives
A reverse mortgage is a loan that converts the borrower’s home equity into
payments from a lender and typically does not require any repayment as
long as the borrower continues to live in the home. Available to
homeowners aged 62 and older, these loans have become an increasingly
popular financial tool for seniors. Almost all reverse mortgages are
currently made under the Home Equity Conversion Mortgage (HECM)
program administered by the Federal Housing Administration (FHA) of the
Department of Housing and Urban Development (HUD). 1 FHA insures
lenders against losses on these mortgages, and charges borrowers
insurance premiums to cover anticipated insurance claims. The number of
HECMs made has grown rapidly in recent years, rising from 157 loans in
fiscal year 1990 to more than 112,000 loans in fiscal year 2008.
The Housing and Economic Recovery Act of 2008 (HERA) made several
modifications to the HECM program. 2 The first of these changes affects
the origination fees that borrowers pay for these loans. Prior to HERA, the
origination fee was 2 percent of the “maximum claim amount”—the lesser
of the home value or the HECM loan limit. HERA changed the fee
calculation to 2 percent of the maximum claim amount up to $200,000 plus
1
For information on consumer protection issues regarding HECMs, see GAO, Product
Complexity and Consumer Protection Issues Underscore Need for Improved Controls
over Counseling for Borrowers, GAO-09-606 (Washington, D.C.: June 29, 2009).
2
P.L. 110-289
Page 1 GAO-09-836 Reverse Mortgages

1 percent of the maximum claim amount over $200,000 with a maximum
fee of $6,000. In conjunction with this change, HUD increased the
minimum origination fee from $2,000 to $2,500 to ensure that lenders
retain an incentive to continue to serve borrowers living in lower-valued
properties. 3 The second of these changes affects the program’s loan limit,
the maximum loan amount that HUD can insure. Specifically, HERA
established a national loan limit for HECMs, which was set at $417,000—a
level substantially higher than the county-based limits that existed prior to
HERA. 4
In light of these changes, HERA contains a mandate for GAO to evaluate

the impact of HERA’s provisions on the availability of credit under the
HECM program, the cost to borrowers participating in the program, and
the program’s financial soundness. As agreed with your offices, this report
examines (1) how HERA’s changes to the HECM program and other
factors have affected HECM lenders’ planned participation in the reverse
mortgage market, (2) the extent to which HERA’s changes to HECM
origination fees and loan limits will affect costs to borrowers and the loan
amounts available to them, and (3) HUD’s actions to evaluate the financial
performance of the HECM program, including the potential impact of loan
limit and house price changes.
To address these objectives, we reviewed laws, regulations, and guidance

relevant to the HECM program, including provisions in HERA and the
American Recovery and Reinvestment Act of 2009 (ARRA). We also spoke
with agency, industry, and nonprofit officials, including those at HUD,
Ginnie Mae, Fannie Mae, AARP, the National Reverse Mortgage Lenders
Association (NRMLA), and the Mortgage Bankers Association (MBA)
about the implications of the HERA changes. In addition, to determine
how HERA’s provisions have affected lenders’ planned participation in the
reverse mortgage market we conducted a survey of a representative
sample of lenders that originated 10 or more HECMs in fiscal year 2008.
The survey included questions about lenders’ plans to offer reverse
mortgages, resources dedicated to their HECM business, consumer
3
The Secretary of HUD has the authority to set a minimum origination fee. For purposes of
this report, our use of the terms “HERA provisions” and “HERA changes” includes HUD’s
change to the program’s minimum origination fee.
4
Prior to HERA, some parts of Hawaii had HECM loan limits exceeding $417,000. HERA did
not change those limits. As a result of the American Recovery and Reinvestment Act of
2009, the HECM loan limit was increased in all areas of the country to $625,500 through
December 31, 2009.

demand for HECMs, and margin rates for HECMs. We received responses
from 57 percent of the lenders we surveyed. In general, the estimates we
made from our survey results had margins of error of plus or minus 10
percentage points at the 95 percent confidence interval. To determine the
extent to which HERA’s changes to HECM origination fees and loan limits
have affected costs to borrowers and the loan amounts available to them,
we reviewed rules for determining borrower costs and loan amounts. We
also obtained and analyzed loan-level data from HUD on HECM loans and
borrowers, which we determined to be reliable for the purposes of this
report. We compared the actual up-front costs and loan funds available to
borrowers who obtained HECMs in 2007 to what their costs and available
loan funds would have been under the HERA provisions. To examine
HUD’s actions to evaluate the financial performance of the HECM
program, we reviewed HUD budget estimates, financial statements, and
actuarial reviews for the HECM program, as well as other analyses the
agency has conducted of program costs. We also reviewed HUD Office of
the Inspector General (OIG) audits of FHA’s financial statements.
Additionally, we reviewed federal agency standards for managing credit
programs, such as those contained in the Federal Credit Reform Act
(FCRA), related Office of Management and Budget requirements and
instructions, and Federal Accounting Standards Advisory Board guidance.
Finally, we interviewed FHA officials, HUD OIG officials, industry
participants, and mortgage market analysts.
We conducted this performance audit from September 2008 through July

2009, in accordance with generally accepted government auditing
standards. Those standards require that we plan and perform the audit to
obtain sufficient, appropriate evidence to provide a reasonable basis for
our findings and conclusions based on our audit objectives. We believe
that the evidence obtained provides a reasonable basis for our findings
and conclusions based on our audit objectives. A more extensive
discussion of our scope and methodology appears in appendix I.
A reverse mortgage is a loan against the borrower’s home that the

Background borrower does not need to repay for as long as the borrower meets certain
conditions. These conditions, among others, require that borrowers live in
the home, pay property taxes and homeowners’ insurance, maintain the
property, and retain the title in the borrower’s name. Reverse mortgages
typically are “rising debt, falling equity” loans, in which the loan balance
increases and the home equity decreases over time. As the borrower
receives payments from the lender, the lender adds the principal and
interest to the loan balance, reducing the homeowner’s equity. This is the

opposite of what happens in forward mortgages, which are characterized
as “falling debt, rising equity” loans. With forward mortgages, monthly loan
payments made to the lender add to the borrower’s home equity and
decrease the loan balance (see fig. 1).
Figure 1: Comparison of 30-Year Forward and Reverse Mortgages
Dollars (in thousands) Traditional 30-year forward mortgage

600
500
400
300
200
100
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Age of loan in years
Dollars (in thousands) Reverse mortgage, over 30 years

600
500
400
300
200
100
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Age of loan in years
Home equity
Loan balance (debt)
Source: GAO.
Note: The graphs are based on a starting house value of $300,000, with an annual 2 percent home
appreciation rate. The interest rates are assumed to be fixed at 5 percent for the forward mortgage
and 3 percent for the reverse mortgage.
There are two primary types of reverse mortgages, HECMs and proprietary
reverse mortgages. The Housing and Community Development Act of 1987
(P.L. 100-242) authorized HUD to insure reverse mortgages and
established the HECM program. According to industry officials, HECMs
account for more than 90 percent of the market for reverse mortgages.
Homeowners aged 62 or older with a significant amount of home equity
are eligible, as long as they live in the house as the principal residence, are
not delinquent on any federal debt, and live in a single-family residence. If

the borrower has any remaining balance on a forward mortgage, this
generally must be paid off first (typically, taken up-front from the reverse
mortgage). In addition, the condition of the house must meet HUD’s
minimum property standards, but a portion of the HECM can be set aside
for required repairs. The borrower makes no monthly payments, and there
are no income or credit requirements to qualify for the mortgage. Lenders
have offered non-HECM, or proprietary, reverse mortgages in the past, but
these products have largely disappeared from the marketplace due, in
part, to the lack of a secondary market for these mortgages. Typically,
proprietary reverse mortgages have had higher loan limits than HECMs but
paid out a lower percentage of the home value to borrowers.
The volume of HECMs made annually has grown from 157 loans in fiscal
year 1990 to more than 112,000 loans in fiscal year 2008. The HECM
program has experienced substantial growth, as the number of HECMs
insured by FHA has nearly tripled since 2005 (see fig. 2).
Figure 2: Number of HECMs Insured Annually, Fiscal Years 1990 through 2008
HECMs (in thousands)
120
100
80
60
40
20
0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Fiscal year
Source: GAO analysis of HECM data.
Additionally, the potential liability of loans insured by FHA has doubled in

the last 2 years (see fig. 3). The potential liability is the sum of the
maximum claim amounts for all active HECMs since the program’s
inception.

Figure 3: Potential Liability of Active HECM Loans, Fiscal Years 2006 through 2008
Dollars (in billions)
90
80
70
60
50
40
30
20
10
0
2006 2007 2008
Fiscal year
Source: GAO analysis of FHA’s financial statements (2006 through 2008).
Finally, recent years have seen a rapid increase in the number of lenders
participating in the HECM program (see fig. 4). However, the bulk of
HECM business is concentrated among a relatively small percentage of
lenders. In fiscal year 2008, roughly 80 percent of all HECMs were
originated by fewer than 300 lenders, or about 10 percent of HECM
lenders.

Figure 4: Number of HUD-Approved Lenders Originating HECMs, Fiscal Years 2003
through 2008
Lenders
3,000
2,500
2,000
1,500
1,000
500
0
2003 2004 2005 2006 2007 2008
Fiscal year
Source: GAO analysis of HECM data.
Lenders can participate in the HECM market through wholesale or retail

channels. Wholesale lenders fund loans originated by other entities,
including mortgage brokers and loan correspondents. Retail lenders
originate, underwrite, and close loans without reliance on brokers or loan
correspondents. Most lenders participate in the HECM market through
retail lending, although some participate through the wholesale process,
and a few have both a retail and wholesale HECM business.
There is a secondary market for HECMs, as most lenders prefer not to

hold the loans on their balance sheets. Fannie Mae has purchased 90
percent of HECM loans and holds them in its portfolio. In 2007, Ginnie
Mae developed and implemented a HECM Mortgage Backed Security
product, in which Ginnie Mae-approved issuers pool and securitize a small
proportion of HECMs. Fannie Mae and Ginnie Mae’s involvement in the
HECM secondary market helps to provide liquidity so that lenders can
continue offering HECM loans to seniors.
The amount of loan funds available to the borrower is determined by

several factors (see fig. 5).

• First, the loan amount is based on the “maximum claim amount,” which is
the highest sum that HUD will pay to a lender for an insurance claim on a
particular property. It is determined by the lesser of the appraised home
value or the HECM loan limit. In the past year, Congress has raised the
HUD loan limit for HECMs twice: HERA established for the first time a
national limit for HECMs, which was set at $417,000. As a result of ARRA,
the national limit was raised again to $625,500 through December 31, 2009.
Prior to HERA, the loan limit for HECMs varied by location and generally
were set at 95 percent of the local area median house price.
• Second, to manage its insurance risk, HUD limits the loan funds available
to the borrower by applying a “principal limit factor” to the maximum
claim amount. 5 HUD developed a principal limit factor table using
assumptions about loan termination rates—which are influenced by
borrower mortality and move-out rates—and long-term house price
appreciation rates, and indexed the table by (1) the borrower’s age and (2)
the expected interest rate—the 10-Year Treasury rate plus the lender’s
margin. The lender determines which factor to use by inputting the
borrower’s current age and the current interest rate information. The older
the borrower, the higher the loan amount; the greater the expected
interest rate of the loan, the smaller the loan amount.
• Third, the funds available to the borrower are further reduced by a

required servicing fee set-aside and by the up-front costs (which include a
mortgage insurance premium and the origination fee), because borrowers
can choose to finance them. HUD allows lenders to charge up to $35 as a
monthly HECM servicing fee. The lender calculates the servicing fee set-
aside by determining the total net present value of the monthly charged
servicing fees that the borrower would pay between loan origination and
when the borrower reaches age 100. 6 The set-aside limits the loan funds
available but is not added to the loan balance at origination. If borrowers
choose to finance up-front costs as part of the loan, the loan funds
available are reduced by these costs.
5
The principal limit factor can range from 20.4 to 90 percent of the maximum claim amount.
6
Present value expresses the worth a future stream of cash inflows and outflows in terms
of an equivalent lump sum received (or paid) today. Net present value is the present value
of estimated future cash inflows minus the present value of estimated future cash outflows.

Figure 5: FHA Process to Determine the Maximum Claim Amount and the Amount of Loan Funds Available
Principal _
Maximum X = Loan Up-front + Servicing = Loan funds
limit costs fee set aside
claim amount amount available
factor
FHA loan
limit Borrower’s 10 yr.
Lesser
value of or age Treasury
rate + Up-front mortgage
House lender insurance premium
value margin + origination fee
Source: GAO.
Borrowers incur various costs when obtaining a HECM. HUD allows

borrowers to finance both up-front and long-term costs through the loan,
which means they are added to the loan balance.
• Origination fee: Prior to HERA, HECM borrowers were charged an

origination fee equal to 2 percent of the maximum claim amount with a
minimum fee of $2,000. Since the implementation of HERA, HECM
borrowers are charged an origination fee calculated as 2 percent of the
maximum claim amount up to $200,000 plus 1 percent of the maximum
claim amount over $200,000, with a maximum fee of $6,000 and a
minimum fee of $2,500.
• Mortgage insurance premium: Borrowers are charged an up-front

mortgage insurance premium equal to 2 percent of the maximum claim
amount. While the maximum claim amount is always higher than the initial
amount a borrower can receive in HECM payments from the lender, FHA
charges the mortgage insurance premium based on this amount because
the loan balance (with accumulated interest and fees) could exceed the
amount a borrower receives in payments and potentially reach the
maximum claim amount. Additionally, borrowers are charged a monthly
mortgage insurance premium on their loan balance at an annual rate of 0.5
percent.
• Interest: Borrowers are charged interest, which generally includes a base

interest rate plus a fixed lender margin rate, on the loan balance. Lenders
can offer HECMs with fixed, annually adjustable, or monthly adjustable
base interest rates. The adjustable rates can be tied to either the 1-Year

Constant Maturity Treasury Rate or 1-Year London Interbank Offered Rate
Index. Most HECMs have adjustable interest rates.
• HECM counseling fee: The HECM program requires prospective

borrowers to receive counseling to ensure an understanding of the loan.
HUD allows counseling providers to charge borrowers up to $125 for
HECM counseling.
• Loan servicing fee: Borrowers pay a monthly servicing fee of up to $35.
• Closing costs: HECMs also have other up-front closing costs, such as
appraisal and title search fees.
FHA’s insurance for HECMs protects borrowers and lenders in four ways.
First, lenders can provide borrowers with higher loan amounts than they
could without the insurance. Second, when the borrower is required to
repay the loan to the lender, if the proceeds from the sale of the home do
not cover the loan balance, FHA will pay the lender the difference. Third,
if the lender is unable to make payments to the borrower, FHA will assume
responsibility for making these payments. Fourth, if the loan balance
reaches 98 percent of the maximum claim amount, the lender may assign
the loan to FHA and FHA will continue making payments to the borrower
if the borrower has remaining funds in a line of credit or still is receiving
monthly payments. To cover expected insurance claims, FHA charges
borrowers insurance premiums, which go into an insurance fund. HECM
loans originated since the inception of the program through 2008 are
supported by FHA’s General Insurance and Special Risk Insurance Fund,
which includes a number of FHA mortgage insurance programs for single-
family and multifamily housing and hospitals. Pursuant to HERA, FHA
moved the HECM program and other insurance programs for single-family
housing into FHA’s Mutual Mortgage Insurance Fund.
FCRA requires federal agencies that provide loan guarantees to estimate

the expected cost of programs by estimating their future performance and
reporting the costs to the government in their annual budgets. 7 Under
credit reform procedures, the cost of loan guarantees, such as mortgage
insurance, is the net present value of all expected cash flows, excluding
administrative costs. This is known as the credit subsidy cost. For loan
guarantees, cash inflows consist primarily of fees and premiums charged
7
FCRA was enacted as part of the Omnibus Budget Reconciliation Act of 1990 (P.L. 101-
508).

to insured borrowers and recoveries on assets, and cash outflows consist
mostly of payments to lenders to cover the cost of claims. Annually,
agencies estimate credit subsidy costs by cohort, or all the loans the
agency is committing to guarantee in a given fiscal year. The credit subsidy
cost can be expressed as a rate. For example, if an agency commits to
guarantee loans totaling $1 million and has estimated that the present
value of cash outflows will exceed the present value of cash inflows by
$15,000, the estimated credit subsidy rate is 1.5 percent. When estimated
cash inflows exceed estimated cash outflows, the program is said to have
a negative credit subsidy rate. When estimated cash outflows exceed
estimated cash inflows, the program is said to have a positive credit
subsidy rate—and therefore requires appropriations.
Generally, agencies are required to produce annual updates of their

subsidy estimates—known as re-estimates—of each cohort based on
information about the actual performance and estimated changes in future
loan performance. This requirement reflects the fact that estimates of
subsidy costs can change over time. Beyond changes in estimation
methodology, each additional year provides more historical data on loan
performance that may influence estimates of the amount and timing of
future claims. Economic assumptions also can change from one year to
the next, including assumptions on home prices and interest rates. FCRA
recognized the difficulty of making subsidy cost estimates that mirrored
actual loan performance and provides permanent and indefinite budget
authority for re-estimates that reflect increased program costs.
In combination, HERA’s changes to the HECM loan limit and origination

Most HECM Lenders fee calculation have had a positive to neutral influence on most lenders’
View the Overall plans to start or continue offering HECMs. Other factors have had varying
influences on lenders’ planned participation. Current economic conditions
Effect of the HERA have had a moderate upward influence on lenders’ plans; however,
Provisions as Neutral secondary market conditions have had a downward influence on about
one-third of lenders’ plans to start or continue offering HECMs. Finally,
or Positive for Their the HERA changes have not influenced most lenders’ plans to offer
Reverse Mortgage proprietary—non-HECM—products.
Business

HERA’s Changes and Other HERA’s changes to the HECM program have had varying effects on HECM
Factors Have Had Varying lenders’ planned participation in the HECM market. On the basis of
Effects on Lenders’ questionnaire responses from a random sample of HECM lenders, we
estimate that for 50 percent of lenders, the combined effect of these
Planned Participation in changes has had an upward influence on their plans to start or continue to
the HECM Market offer HECMs (see fig. 6). 8 For 42 percent of lenders, the combination of
HERA’s changes to the origination fee and loan limits for the HECM
program have had little to no influence on their plans to offer HECMs,
while for 8 percent of lenders, HERA’s changes have had a downward
influence. Some industry participants we interviewed stated that the
changes were a good compromise that benefited borrowers by limiting the
origination fee and increasing the loan limit, thereby increasing the money
borrowers could receive from a HECM. Additionally, officials at NRMLA
and MBA said the changes benefited lenders by making the product more
attractive to individuals with higher-value homes.
8
We surveyed a random sample of the 2,779 lenders that originated HECMs on a retail basis
in fiscal year 2008. For purposes of this report, we define retail lenders as lenders that
originate HECMs as opposed to funding HECMs originated by other lenders. For our survey
questions about HERA’s changes and other factors influencing lenders’ planned
participation, we asked lenders the following: “How, if at all, has (factor x) influenced your
institution’s likelihood to start or continue to offer HECMs on a retail basis?” Accordingly,
our results apply only to lenders’ retail HECM business. Unless otherwise noted, our
estimates have margins of error of plus or minus 10 percentage points or less at the 95
percent confidence interval. See appendix I for additional information on this survey
methodology.

Figure 6: Influence of HERA’s Provisions on Loan Limits and Fees and Other Factors on Lenders’ Plans to Offer HECMs
HERA’s provisions on loan Moderate or great Little or no Great to moderate Don’t Not
limits and fees upward influence influence downward influence know applicable
HERA's changes to the

11% 65 % 22 % 2% N/A %
calculation of origination fees
HERA's changes to
70 28 1 1 N/A
HECM loan limits
HERA's changes to both HECM

50 42 8 1 N/A
origination fees and loan limits
Other factors
Implementation of the
67 29 1 2 0
HECM for Purchase program
Current economic conditions 52 28 19 0 0
Reduced opportunities in the

forward mortgage market 34 54 8 0 4
HERA's prohibition of participation

of non-FHA approved entities in 33 48 13 4 1
the origination of HECMs
Current secondary market options 14 50 32 0 4
Current house price trends 10 50 38 1 1
HERA's prohibition of 9 60 29 1 0
lender-funded HECM counseling
Current availability of
9 56 22 1 12
wholesale lending partners
HERA's restrictions on selling

other financial products in 5 78 7 6 4
conjunction with HECMs
Source: GAO analysis of survey of HECM lenders.
Note: Figure shows estimates based on GAO survey of HECM lenders. Estimates have margins of
error of plus or minus 10 percentage points or less at the 95 percent confidence interval.
Taken separately, the two HERA provisions have had differing effects on
lenders’ plans to offer HECMs. We estimate that for about 70 percent of
lenders, HERA’s increase in HECM loan limits has had an upward

influence on the likelihood of offering HECMs. 9 The loan limit increase has
had little to no influence on almost all of the remaining lenders’ plans to
offer HECMs. We estimate that 86 percent of lenders expect that HERA’s
creation of a single national loan limit of $417,000 will somewhat or greatly
increase consumer demand for HECMs.
Although the increase in the loan limit has generally had an upward
influence on lenders’ plans, the change to the calculation of the origination
fee has had a different effect. We estimate that changing how the fee is
calculated has had a downward influence on plans to offer HECMs for 22
percent of HECM lenders, little to no influence for 65 percent of lenders,
and an upward influence for 11 percent of lenders. Consistent with these
views, 65 percent of lenders expect the change in origination fee to have
no effect on consumer demand for HECMs. An estimated 26 percent of
lenders expect the change in the origination fee to increase consumer
demand, while only a few lenders expect the change to decrease consumer
demand.
We estimate that only 2 percent of HECM lenders do not plan to continue

to offer HECMs. Of the respondents in our sample, three lenders indicated
that they did not plan to continue offering HECMs. None of these were
large HECM lenders, as they each originated from 40 to 160 HECMs in
fiscal year 2008. Each of these lenders participated in the HECM market
solely through their retail business. These three lenders varied in the
amount of time that they have offered the HECM product. A representative
of one lender indicated that HERA’s changes to the loan limits and
origination fee had a great upward influence on the likelihood that it
would offer HECMs, but nonetheless planned to discontinue offering
HECMs. The other two lenders indicated that HERA and other economic
factors had little to no influence on their decision to discontinue offering
HECMs, and one of these lenders noted on the survey that it had
discontinued offering HECMs before the enactment of the HERA.
As part of our survey, we asked lenders how various economic and

legislative factors influenced their plans to start or continue offering
HECMs. Two factors had an upward influence on most lenders’ plans to
offer HECMs in 2009. For an estimated 67 percent of HECM lenders, the
implementation of the HECM for Purchase program (authorized by HERA)
9
See appendix II for survey results pertaining to the ARRA’s increase in the HECM loan
limit to $625,500.

has had an upward influence on their plans to offer HECMs, and it has had
little to no influence on almost all of the remaining lenders’ HECM
origination plans. 10 Some industry participants told us that the HECM for
Purchase program likely will make HECMs attractive to a broader range of
seniors. Additionally, current economic conditions have had an upward
influence on the plans to offer HECMs for about 52 percent of lenders.
NRMLA officials explained that seniors are seeking additional revenue
because they have less available income from traditional sources, such as
interest and dividend payments and retirement accounts, which is partially
attributable to poor economic and financial market conditions.
Additionally, two other factors have had an upward influence on some
lenders’ plans to offer HECMS. For about one-third of lenders, both (1)
reduced opportunities in the forward mortgage market and (2) HERA’s
prohibition on the participation of non-FHA approved entities in the
origination of HECMs has had a moderate or great upward influence on
their plans to offer HECMs. 11
In contrast, three factors had more of a downward influence on some

lenders’ planned participation in the HECM market. First, we estimate
from our survey that house price trends have had a downward influence
on the HECM origination plans of 38 percent of lenders; however, house
price trends had little or no influence on plans for about 50 percent of
lenders. Some industry participants told us that the recent decline in
house prices has prevented some seniors from obtaining a HECM either
because they lack the equity in their home to qualify for the loan, or
because they would not receive enough funds from the HECM to have any
cash remaining after they deduct HECM fees and pay off any existing
mortgage debt.
Second, we estimate that the availability of secondary market options has

had a downward influence on the plans of about one-third of lenders to
offer HECMs. The secondary market for HECMs plays an important role in
10
HERA authorized a HECM for Purchase program for seniors who wish to use a HECM to
buy a new home. Unlike a traditional HECM, a HECM for purchase is made against the
value of the home to be purchased rather than against the value of a home the borrower
already owns. The HECM for purchase program allows a senior to simultaneously buy a
new home and obtain a HECM in a single transaction with a single set of closing costs,
reducing the cost to the senior.
11
HERA requires that all parties participating in the origination of HECMs be approved by
HUD. This prohibits non-HUD-approved mortgage brokers—sometimes called HECM
advisors—from offering HECMs.

maintaining availability of loans because lenders prefer not to hold
HECMs on their balance sheets. There are currently two primary options
in the secondary market—Fannie Mae and Ginnie Mae.
• Fannie Mae officials stated that Fannie Mae bought and held more than 90
percent of HECMs in its portfolio in 2008 and was the principal secondary
market purchaser of HECM loans. However, Fannie Mae’s regulator—the
Federal Housing Finance Agency—recently required it to reduce the
mortgage assets it holds in portfolio. Fannie Mae officials told us that as a
result, they are making changes to their HECM business, which will attract
other investors to the secondary market for HECMs, in order to decrease
their share of the market. Recently, Fannie Mae lowered the price it pays
lenders for HECMs and implemented a “live pricing” system that requires
lenders to commit to the volume of HECMs they will sell to Fannie Mae. 12
We estimate that approximately 90 percent of lenders viewed secondary
market pricing requirements and the transition to live pricing as important
factors in recent margin rate increases on HECMs. Fannie officials
explained that as the price they pay lenders for HECMs falls, the margin
rate the lenders charge the consumers generally increases. 13 Some lenders
we surveyed noted that margin rate increases stemming from pricing
changes could make HECMs less attractive to borrowers because they
would not be able to obtain as much cash from their HECM. 14 Some
lenders noted that live pricing complicates their relationship with
borrowers because the interest rate can change between loan application
and closing, which may result in the senior being able to receive less
money from their HECM than originally quoted.
12
Live pricing eliminated the predetermined pricing contracts that were previously used, in
which lenders negotiated with Fannie Mae to set the price of all loans made within a
predetermined period. Instead, live pricing implemented a process whereby lenders enter
into a commitment with Fannie Mae to sell a specified volume of loans at a specified price
within 90 days. Fannie Mae officials noted that this system makes HECM pricing more
similar to the pricing system for forward mortgages in the secondary market. According to
some industry observers, some lenders are reluctant to guarantee the interest rate on a
HECM well in advance of closing because it can take several months to close a HECM and
the loan may not close before the end of the 90-day commitment period, in which case the
lender would incur a penalty.
13
Fannie Mae maintains a cap on margin rates for HECMs they purchase.
14
As previously discussed, the amount of loan funds available to the borrower depends in
part on the interest rate, which includes the lender’s margin. In general, the higher the
interest rate, the less money a borrower will have available from the HECM.

• Ginnie Mae developed and guarantees a HECM Mortgage Backed Security
(HMBS) that aims to expand the availability of HECMs from multiple
lenders, reduce borrowing costs, and create a broader secondary market
for HECM loans. Ginnie Mae officials stated that they were poised to take
on extra volume in the HECM secondary market by guaranteeing
securities issued by lenders. AARP officials noted that Ginnie Mae’s HMBS
product could help introduce competition into the secondary market for
reverse mortgages, lowering margin rates for seniors. However, industry
participants point to several issues with the Ginnie Mae product that could
limit its appeal to lenders. First, Ginnie Mae requires HMBS issuers to buy
back the HECM when the loan balance reaches 98 percent of the loan’s
maximum claim amount. 15 Second, issuers are required to pay interest
shortfalls to investors when the loan is terminated mid-month. Some
HECM lenders have noted that both of these provisions expose them to
extra risk on the loan, as compared to the alternative of selling the HECM
outright as they had when selling to Fannie Mae.
Third, for an estimated 29 percent of lenders, HERA’s prohibition on

lender-funded counseling has had a downward influence on plans to offer
HECMs. Industry participants said that this prohibition is a problem for
the HECM industry because counseling is required for borrowers to obtain
a HECM, but borrower-paid counseling can be a deterrent for seniors who
are still deciding if they want a HECM, or for those who have limited
financial means to pay for counseling. In contrast to these comments, we
estimate that the prohibition on lender-funded counseling had little or no
influence on the plans of 60 percent of lenders.
Our survey of HECM lenders asked about two other factors—HERA’s

restrictions on selling other financial products in conjunction with HECMs
and the current availability of wholesale lending partners—that could
influence lenders’ plans to start or continue to offer HECMs. In general,
these factors had little or no influence on lenders’ plans (see fig. 6).
HERA Has Not Influenced In 2008, several non-HECM reverse mortgages—referred to as jumbo or
Most Lenders’ Plans to proprietary reverse mortgages—were available in the marketplace.
Offer Non-HECM Reverse Proprietary reverse mortgages offered loan limits that were greater than
the HECM loan limit. For example, Financial Freedom, a large reverse
Mortgages mortgage lender, offered a product called the Cash Account Advantage
Plan, which was not subject to the HECM loan limits, and in some cases
15
Generally, issuers can then assign the loan to HUD.

provided more cash than a HECM to borrowers with higher-value homes.
Based on our survey results, we estimate that approximately 43 percent of
HECM lenders made non-HECM reverse mortgages in 2008. However,
towards the end of 2008, almost all of the non-HECM reverse mortgage
products were withdrawn from the market due to the lack of a secondary
market to support them. Nonetheless, from our survey results, we estimate
that 36 percent of HECM lenders plan to offer a non-HECM reverse
mortgage in 2009.
We estimate that HERA’s changes to the calculation of the origination fee

and loan limit have had little or no influence on 68 percent of lenders’
plans to originate non-HECM reverse mortgages (see fig. 7). However, for
an estimated 29 percent of HECM lenders, HERA’s change to the loan
limits has had an upward influence on their plans to offer non-HECM
reverse mortgages. Additionally, we estimate that for 32 percent of
lenders, the implementation of the HECM for Purchase program had an
upward influence on their plans to offer these loans. We estimate that
current economic conditions have had an upward influence on plans to
offer non-HECM reverse mortgages for 29 percent of lenders, little to no
influence for 34 percent of lenders, and a downward influence for 17
percent of lenders. Our survey of HECM lenders asked about several other
factors (see fig. 7) that could influence lenders’ plans to offer a non-HECM
reverse mortgage product in 2009. Generally, these factors have had little
or no influence on lenders’ plans. Our survey results did not indicate that
secondary market conditions had a downward influence on the plans of
most lenders. However, several lenders we interviewed said that while
they hoped to offer a non-HECM reverse mortgage in 2009, their ability to
do so would depend on the availability of funding in the secondary market.

Figure 7: Influence of HERA’s Provisions on Loan Limits and Fees and Other Factors on Lenders’ Plans to Offer Non-HECM
Reverse Mortgages in 2009
HERA’s provisions on loan Moderate or great Little or no Great to moderate Don’t Not
limits and fees upward influence influence downward influence know applicable
HERA's changes to the

9% 74 % 12 % 5% N/A %
calculation of origination fees
HERA's changes to
29 56 10 6 N/A
HECM loan limits
HERA's changes to both HECM

15 68 12 5 N/A
origination fees and loan limits
Other factors
Implementation of the 32 43 5 4 16
HECM for Purchase program
Current economic conditions 29 34 17 3 16
Current secondary market options 19 38 22 5 17
Reduced opportunities in the 17 53 7 4 19

forward mortgage market
HERA's prohibition of participation

of non-FHA approved entities in 16 57 6 5 17
the origination of HECMs
Current availability of
14 45 21 3 16
wholesale lending partners
Current house price trends 12 45 24 3 16
HERA's prohibition of
8 61 12 4 16
lender-funded HECM counseling
HERA's restrictions on selling

other financial products in 5 71 3 6 15
conjunction with HECMs

HERA’s provisions will affect borrowers in varying ways depending
HERA Provisions Will primarily on home value and whether HERA’s increase in loan limit will
Affect Borrower change the maximum claim amount of the loan. HERA’s changes to HECM
origination fees and loan limits are likely to change the up-front costs
Costs and Loan (origination fee and up-front mortgage insurance premium) and the loan
Amounts Differently funds available for most new borrowers. Our analysis of data on
borrowers who took out HECMs in 2007 shows that had the HERA
Depending on Home provisions been in place, most borrowers would have paid less or the
Value and Other same amount in up-front costs, and most would have had more or the
Factors same amount of loan funds available. Additionally, about 28 percent of
HECM borrowers in 2007 would have seen an increase in maximum claim
amount due to HERA’s increase in loan limit, which would have meant
more loan funds available for nearly all of these borrowers. Borrowers
also may be affected by other consequences of the HERA provisions, such
as margin rate increases and changes to funding of HECM counseling.
HERA Provisions Will The net effect of the HERA provisions on an individual borrower’s total
Change Up-front Costs for up-front costs depends on house value, the local loan limit prior to HERA,
Many Borrowers and the new loan limit. HECM up-front costs consist primarily of the up-
front mortgage insurance premium and the origination fee, both of which
are calculated as a proportion of the maximum claim amount. Most
borrowers are likely to see changes in origination fees due to HERA.
Generally, those with house values greater than the prior HECM loan limit
in their area will see changes in the up-front mortgage insurance
premium. 16 Borrowers fall into two categories, based on whether their
maximum claim amount changes:
• Maximum claim amount does not change: For borrowers whose houses
are valued at or less than the prior HECM loan limit in their area, the
maximum claim amount does not change. Therefore, for these borrowers,
the mortgage insurance premium (which is calculated based on the
maximum claim amount) also does not change. However, the origination
fee may change depending on the value of the house. A borrower whose
house is valued at less than $125,000 should expect up to a $500 increase
in the up-front costs due to the increase in the minimum origination fee
from $2,000 to $2,500. A borrower whose house is valued at $125,000 to
16
HERA did not change the HECM loan limits in parts of Hawaii that had loan limits higher
than $417,000 prior to HERA. As a result, borrowers in those areas whose house values are
greater than the pre-HERA loan limits will see no change in their up-front mortgage
insurance premiums due to HERA.

$200,000 would see no change in the up-front costs because they would
pay the same 2 percent of the maximum claim amount (the same as before
HERA). A borrower whose house is valued at greater than $200,000 would
expect a decrease in up-front costs due to the decreased origination fee
for amounts greater than $200,000 and the fee cap of $6,000. For an
example, see borrower D, whose house value is $300,000, in table 1.
• Maximum claim amount increases: For borrowers whose maximum

claim amount increases because their house values are greater than the
prior local HECM loan limit, the change to up-front costs is more complex.
All borrowers in this category will pay more in up-front mortgage
insurance premiums because premiums are calculated based on the entire
maximum claim amount. However, some borrowers may pay more in
origination fees, while others will pay less. When combining these two
costs, the total up-front costs could increase, decrease or remain the same.
For example, borrowers A, B, and C in table 1 each own houses valued at
$300,000 that are located in counties in which prior HECM loan limits
varied from $200,000 to $290,000. Each borrower would see different
effects in up-front costs.
See appendix III for a more complete explanation of how up-front costs
will change for borrowers with different characteristics.
Table 1: Change in Up-front Costs for $300,000 Houses in Various Locations
Up-front
Prior local mortgage
HECM loan Maximum claim insurance Origination Total up- Change in up-
Borrower limit amount premium fee front costs front costs
A $200,000 Before HERA $200,000 $4,000 $4,000 $8,000 $3,000 more
After HERA 300,000 6,000 5,000 11,000
B 275,000 Before HERA 275,000 5,500 5,500 11,000 No change
After HERA 300,000 6,000 5,000 11,000
C 290,000 Before HERA 290,000 5,800 5,800 11,600 $600 less
After HERA 300,000 6,000 5,000 11,000
D 325,000 Before HERA 300,000 6,000 6,000 12,000 $1,000 less
After HERA 300,000 6,000 5,000 11,000
Source: GAO.

Most 2007 HECM To illustrate the potential effect of the HERA provisions on borrowers, we
Borrowers Would Have compared the actual maximum claim amounts, up-front costs (origination
Paid the Same or Less in fee plus the up-front insurance premium), and loan funds available for
HECM borrowers in 2007 to what their maximum claim amounts, up-front
Up-front Costs under the costs, and loan funds available would have been had the HERA provisions
HERA Provisions, and been in place. 17 Overall, we found that nearly 27 percent of borrowers
Most Borrowers Would would have paid more in up-front costs, 46 percent would have paid less,
Have Had the Same or and 27 percent would have paid the same (see fig. 8). The amount and
More Loan Funds direction of the changes to up-front costs and loan funds available
Available primarily depended on house value and whether a borrower would have
benefited from an increase in loan limit (about 28 percent of 2007 HECM
borrowers’ homes were valued at more than the prior loan limit and would
have seen their maximum claim amounts increase because of HERA’s
increase in the loan limit).
Our analysis of up-front costs broken down by its two components is as

follows:
• Origination fees: About 24 percent of 2007 borrowers would have paid

more in origination fees, 49 percent would have paid less, and 27 percent
would have paid the same amount. Increases in origination fees were due
either to the $500 increase in the minimum origination fee (about 17
percent of all borrowers) or to the increased loan limits (about 6 percent
of all borrowers). Borrowers who would have paid less in origination fees
had maximum claim amounts greater than $200,000, which means they
would have benefited from the decrease in the origination fee for the
portion of the maximum claim amount greater than $200,000, the $6,000
origination fee cap, or both.
• Up-front mortgage insurance premium: Twenty-eight percent of 2007

HECM borrowers would have paid more in up-front mortgage insurance
premiums due to increases in the loan limit, while 72 percent of borrowers
would have paid the same amount, generally because the size of their
loans was limited by the value of their homes and not the HECM loan
limit. 18
17
Loan funds available for a HECM, is the loan amount after applying the principal limit
factor to the maximum claim amount and deducting the up-front costs and servicing fee
set-aside (see fig. 5).
18
For some Hawaii borrowers, the up-front mortgage insurance premium would have been
the same because their loans were capped by loan limits that did not change due to HERA.

Changes in the loan limits and up-front fees would have affected the loan
funds available to most 2007 borrowers. Borrowers whose maximum claim
amount would have increased because of an increase in loan limit would
have paid a higher up-front mortgage insurance premium, regardless of
how much of their available loan funds they chose to access. Because this
analysis assumed that HECM borrowers financed the up-front costs in the
loan, any increase or decrease in the up-front costs affects the amount of
loan funds that are available to them. Our analysis—which assumes that
borrowers financed their up-front costs—shows that had the HERA
provisions been in place at origination for 2007 HECMs, approximately 56
percent of borrowers would have had more loan funds available, 17
percent would have had less loan funds available, and 27 percent would
have had the same amount available (see fig. 8).
Specifically,
• 28 percent of borrowers would have had more loan funds available,

primarily due to the increase in loan limit;
• about 28 percent of borrowers would have had more loan funds available
due solely to a decrease in their up-front fees;
• 17 percent of borrowers would have had a smaller amount of loan funds

available due solely to an increase in their up-front fees; and
• 27 percent of borrowers would have experienced no change in the amount

of loan funds available because their up-front fees and loan limits
remained the same.

Figure 8: Number of 2007 HECM Borrowers Affected by HERA Provisions
Change in loan funds available

Increase No change
Decrease Total
Change in Increase 9,761 0 17,303 27,064
up-front costs (10%) (17%) (27%)
6
No change 27,548 0 27,554
(less than
(27%) (27%)
1%)
a
46,862 0 0 46,862
(46%) (46%)
Decrease
Total 56,629 27,548 17,303 101,480

(56%) (27%) (17%) (100%)
Source: GAO.
a
For 18,847 of these borrowers, the maximum claim amount increased. For the other 28,015
borrowers, the maximum claim amount remained the same.
Additionally, figure 8 shows the number of 2007 borrowers within the

various categories and figure 9 shows the average changes in up-front
costs and loan funds available for each category of borrower. Borrowers
with the largest increases in their maximum claim amounts on average
would have the largest percent increases in up-front costs (see fig. 9).
Borrowers with no increase in their maximum claim amount, who have a
change to up-front costs, will have a corresponding change in loan funds
available that are equal in size but opposite in direction. For example a
borrower with a $200 decrease in up-front costs will have a $200 increase
in loan funds available and a borrower with a $300 increase in up-front
costs will have a $300 decrease in loan funds available.

Figure 9: Average Changes in Maximum Claim Amounts, Up-front Costs, and Loan
Funds Available for 2007 HECM Borrowersa
Average Average Average

Change Change in original increase in Average difference in
in up-front loan funds Number of maximum maximum difference in loan funds
costs available 2007 HECMs claim amount claim amount up-front costs available
$1,508 $42,560
9,761 $240,008 $64,368 (11%) (28%)
402 -402
17,303 90,760 0 (5%) (-1%)
0 14,232
6 260,875 20,292 (0%) (8%)
0 0
27,548 162,592 0 (0%) (0%)
-422 27,883
18,847 b 350,401 39,477 (-2%) (12%)
733
c -733
28,015 270,918 0 (less than
(-4%)
1%)
Increase
Decrease
No change
Source: GAO.
a
This analysis did not include refinanced HECMs, but a separate analysis of refinanced HECMs
showed similar changes in maximum claim amounts, up-front costs, and loan funds available within
the different categories of borrowers.
b
For these borrowers, the maximum claim amount increased.
c
For these borrowers, the maximum claim amount remained the same.
Borrowers May Be Increased lender margin rates stemming from HERA’s change to the
Affected by Other Factors, origination fee calculation could reduce loan funds available to borrowers.
Such as Lender Margin At loan origination, the expected interest rate HUD uses to determine the
portion of the maximum claim amount that will be made available to the
Rates and Counseling Fees borrower includes the 10-year Treasury rate plus the fixed lender margin
rate. Our survey of HECM lenders indicates that some lenders have raised
their margin rates modestly to compensate for HERA’s limitations on the
origination fee; however, we did not receive a sufficient number of
responses to reliably estimate the median increase in margin rate for the

population. 19 To illustrate the impact of a modest increase in margin-rate
on borrowers, we applied a 0.25 percentage point increase to borrowers
who took out HECMs in 2007. We found that these borrowers would have
seen a 3 percent average decrease in loan funds available as a result of the
higher margin rate. 20 A comparison of HUD data on HECMs originated
within the first 3 months of HERA’s implementation with data from the
same 3 months from the prior year indicates that average margin rates
were higher after HERA but that the overall average HECM expected
interest rates were essentially the same. This outcome resulted from
declines in 10-year Treasury rates offsetting increases in lender margin
rates.
In addition, more borrowers, as well as prospective borrowers who

ultimately do not obtain a HECM, may need to pay counseling fees.
Provisions in HERA prohibit lenders from paying for this counseling but
allow HUD to use a portion of HECM mortgage insurance premiums for
this purpose. HUD officials said that they have not exercised this authority
because the resulting reduction in premium income would affect the
subsidy rate of the program adversely and potentially require
appropriations. Because HUD did not implement this provision, more
borrowers and prospective borrowers may need to pay counseling fees
themselves. For borrowers who do eventually obtain a HECM, the fee can
be financed in the loan. Prospective borrowers who do not qualify for a
HECM or who choose not to proceed with the loan after counseling may
have to pay for counseling out of pocket. HUD’s recent announcement that
it will provide approximately $8 million in grant funds for HECM
counseling in 2009 may mitigate any negative impact the HERA changes
may have on seniors’ ability to obtain HECM counseling.
19
Of the lenders that responded to the survey question, we estimate that 48 percent
increased margin rates for HECMs offered by their institution by at least 0.25 percentage
points. This estimate has a margin of error of within plus or minus 13 percentage points.
20
All other things being equal, an increase in margin rate causes a decrease in loan funds
available (see fig. 5). This analysis uses the 10-year Treasury rate that was available to the
borrower at loan origination.

HUD has taken or planned steps to enhance its analysis of the HECM
HUD Has Enhanced program’s financial performance. However, HUD’s recent estimates of
Its Analysis of HECM program costs indicate weaker performance than previously estimated,
primarily due to more pessimistic assumptions about long-term house
Program Costs but price trends. Additionally, higher loan limits enacted under HERA and the
Changes in House American Recovery and Reinvestment Act of 2009 (ARRA) could increase
HUD’s financial risk.
Price Trends and
Higher Loan Limits
Have Increased HUD’s
Risk of Losses
HUD Is Taking Steps to To estimate the cost of the HECM program, HUD uses a model to project
Improve its Analysis of the the cash inflows (such as insurance premiums paid by borrowers) and
HECM Program’s Financial cash outflows (such as claim payments to lenders) for all loans over their
expected duration. HUD’s model is a computer-based spreadsheet that
Performance incorporates assumptions based on historical and projected data to
estimate the amount and timing of insurance claims, subsequent
recoveries from these claims, and premiums and fees paid by borrowers.
These assumptions include estimates of house price appreciation, interest
rates, average loan size, and the growth of unpaid loan balances. HUD
inputs its estimated cash flows into OMB’s credit subsidy calculator,
which calculates the present value of the cash flows and produces the
official credit subsidy rate for a particular loan cohort. A positive credit
subsidy rate means that the present value of the cohort’s expected cash
outflows is greater than the inflows, and a negative credit subsidy rate
means that the present value of the cohort’s expected cash inflows is
greater than the outflows. To budget for a positive subsidy an agency must
receive an appropriation. HUD also uses the cash flow model to annually
estimate the liability for loan guarantees (LLG), which represents the net
present value of future cash flows for active loans, taking into account the
prior performance of those loans. HUD estimates the LLG for individual
cohorts as well as for all cohorts combined. 21 The LLG is a useful statistic
because unusual fluctuations in the LLG can alert managers to financial
risks that require further attention.
21
The LLG does not include cash flows that have already occurred.

HUD in recent years has enhanced its cash flow model for the HECM
program. In 2007, the HUD Office of Inspector General’s (OIG) annual
audit of FHA’s financial statements cited a material weakness in the cash
flow model FHA used to generate credit subsidy estimates for the HECM
program. 22 Among other things, the audit noted technical errors in the
model, significant discrepancies between projected and actual cash flows,
and a lack of supporting documentation for certain modeling decisions.
Partly in response to the OIG audit, HUD made a number of improvements
to both the model and its supporting documentation, and in 2008 the HUD
OIG eliminated the material weakness. 23 For example, HUD improved the
methodology it uses for its cash flow model. In the past, HUD used
historical averages for termination and recovery rates for projecting cash
flows. In 2008, HUD began to incorporate forecasts of national house price
appreciation and interest rates from IHS Global Insight, an independent
source for economic and financial forecasts, into its modeling. 24
Additionally, HUD improved the way it estimates the growth of unpaid
principal balances, which HUD uses to calculate the LLG. In the past, HUD
used both active and terminated loans to generate this estimate. Since
2008, HUD has included only active loans to generate this estimate, which
is more appropriate because the LLG represents the expected future cash
flows of currently active loans. HUD also developed a master database of
loan-level information to support the HECM cash flow model. Previously,
HUD staff had to draw on data from multiple sources, which increased the
chance of analytical errors. Finally, HUD made a number of enhancements
to its documentation of estimation processes, including how
macroeconomic projections are incorporated into the cash flow model.
HUD plans to subject the HECM program to an annual actuarial review,

which should provide additional insight into the program’s financial
condition. Such a review would likely assess if program reserves and
funding were sufficient to cover estimated future losses, as well as the
sensitivity of this analysis to different economic and policy assumptions.
Historically, the HECM program has not had a routine actuarial review
because it was supported by the General Insurance and Special Risk
22
HUD OIG, Audit of the Federal Housing Administration’s Financial Statements for
Fiscal Years 2007 and 2006, 2008-FO-0002 (Washington, D.C., Nov. 8, 2007).
23
HUD OIG, Audit of the Federal Housing Administration’s Financial Statements for
Fiscal Years 2008 and 2007, 2009-FO-0002 (Washington, D.C., Nov. 7, 2008).
24
IHS Global Insight is a private company that forecasts a wide range of financial and
economic indicators.

Insurance Fund (GI/SRI) Fund, which does not have such a review
requirement. 25 However, as of fiscal year 2009, the HECM program is in the
Mutual Mortgage Insurance (MMI) Fund, which is statutorily required to
receive an independent actuarial review each year and includes FHA’s
largest mortgage insurance program. 26 HUD officials told us that future
actuarial reviews of the MMI Fund will include a separate assessment of
the HECM program.
HUD also is considering producing credit subsidy re-estimates for the

HECM program. 27 As discussed later in this report, HUD has generated
credit subsidy estimates for individual HECM cohorts for several years.
However, HUD officials told us that, until recently, they did not have the
data necessary to produce subsidy re-estimates for HECMs. Specifically,
the officials noted that for HECM cohorts prior to 2009, assets for HECMs
were aggregated with assets from other programs in the GI/SRI Fund and
not accounted for separately. HUD officials said that they are now
accounting for HECM assets separately, which will enable them to
produce re-estimates for the HECM program. Re-estimates can highlight
cohorts that are not expected to meet original budget estimates. This
information could help inform future actions to manage HUD’s insurance
risk and control program costs.
Prior Cost Estimates HUD’s most recent estimates of two important financial indicators for the
Indicated That the HECM HECM program—the credit subsidy rate and the LLG—suggest weaker
Program Was Profitable financial performance than previously estimated, largely due to more
pessimistic house price assumptions. All other things being equal, lower
but Current Estimates house price appreciation can increase HUD’s insurance losses because it
Forecast Losses, Primarily makes it less likely that the value of the home will cover the loan balance.
Due to Revised House Analyses by HUD have found that the financial performance of the HECM
Price Assumptions program is sensitive to long-term trends in house prices. HUD officials told
us that HECM program performance is less sensitive to short-term price
25
Pursuant to congressional directives, HUD submitted reports on the HECM program in
1995, 2000, and 2003 (P.L. 100-242 and P.L. 106-569). These reports included actuarial
reviews of the program.
26
12 U.S.C. Sec. 1708(a)(4). HECMs originated in fiscal year 2009 and beyond will be
accounted for under the MMI Fund. Loans originated in or before fiscal year 2008 will be
accounted for under the GI/SRI Fund.
27
Currently, HUD includes HECMs in its credit subsidy re-estimates for the GI/SRI Fund as
a whole.

declines because borrowers with HECMs, unlike those with traditional
forward mortgages, do not have an incentive to terminate (or default on)
their loans when prices fall.
HUD has made credit subsidy estimates for HECM cohorts from 2006
forward. Because the HECM program was relatively small prior to 2006,
HUD did not produce separate subsidy estimates for the HECM program
but included HECMs in its estimates of subsidy costs for the GI/SRI Fund
as a whole. For the 2006 through 2009 HECM cohorts, HUD estimated
negative subsidy rates ranging from - 2.82 percent in 2007 to -1.37 percent
in 2009 (see fig. 10). However, for the 2010 cohort, HUD estimated a
positive subsidy rate of 2.66 percent. Because HUD is expecting to insure
about $30 billion in HECMs in 2010, this rate corresponds to a subsidy cost
of $798 million. As required by the Federal Credit Reform Act, the
President’s budget for fiscal year 2010 includes a request for this amount. 28
28
On July 17, 2009, the House Committee on Appropriations approved a fiscal year 2010
appropriations bill for HUD. The bill contained a provision directing HUD to adjust, as
necessary, the principal limit factor for new loans to ensure that the program operates
without a subsidy. The provision also prohibited HUD from reducing a borrower’s principal
limit factor below 60 percent. If enacted, this provision would eliminate the need for the
$798 million budget request but would also reduce the loan funds available to some
borrowers.

Figure 10: HECM Credit Subsidy Rates, Fiscal Years 2006 through 2010
Subsidy rate (percentage)

3
2.66%
1
Yes
0 Subsidy required?
No
-1
-1.37%
-2 -1.74% -1.90%
-3 -2.82%
2006 2007 2008 2009 2010
Cohort
Source: GAO analysis of OMB federal credit supplement data (2006 through 2010).
HUD officials told us that the positive subsidy rate for fiscal year 2010
largely was due to incorporating more conservative assumptions about
long-term house price trends than had been used for prior cohorts. For
budgeting purposes, the Administration decided to use more modest
appreciation rates than the private sector forecasts HUD typically uses.
Specifically, the house price appreciation rates used were 0.5 percent
greater than the forecasted inflation rates. HUD officials told us that if
they had used IHS Global Insight projections to develop the fiscal year
2010 credit subsidy estimate, there would be no need for an appropriation
because the credit subsidy rate would be negative.
HUD also has estimated the LLG for the HECM program since 2006. As
shown in figure 11, HUD’s original LLG estimates grew substantially from
2007 to 2008, increasing from $326 million to $1.52 billion. According to
FHA’s financial statements for fiscal years 2007 and 2008, the increase was
primarily due to the lower house price appreciation projections used in the
2008 analysis. 29 The report noted that lower appreciation rates result in
29
HUD, FHA Annual Management Report Fiscal Year 2007 (Washington, D.C., 2007) and
FHA Annual Management Report Fiscal Year 2008 (Washington, D.C., November 2008).

lower recoveries on mortgages assigned to HUD, which in turn increases
HUD’s liability.
Figure 11: Loan Liability Guarantee for the HECM Program, Fiscal Years 2006
through 2008
Dollars (in millions)

1,600
1,400
1,200
1,000
800
600
400
200
0
2006 2007 2008
Fiscal year
Source: GAO analysis of FHA financial statements (2006 through 2008).
In September 2008, HUD analyzed the sensitivity of the 2008 LLG estimate
for the HECM program as a whole to different assumptions, including
alternative house price scenarios. HUD examined the impact of house
price appreciation that was 10 percent higher and 10 percent lower than
the baseline assumptions from IHS Global Insight for fiscal years 2009
through 2013. (For example, for a baseline assumption of 4 percent house
price appreciation, the lower and higher scenarios would have been 3.6
percent and 4.4 percent, respectively.) HUD estimated that the more
pessimistic assumption increased the LLG from $1.52 billion to $1.78
billion, while the more optimistic assumption reduced the LLG to $1.27
billion.

HUD Uses a Conservative When estimating future costs for all HECMS, HUD assumes that the
Approach in Estimating property value at loan origination is equal to the maximum claim amount.
Program Costs, but Higher For loans in which the property value is more than the HECM loan limit,
this approach results in a conservative assumption about the amount of
Loan Limits May Increase home equity available at the end of the loan to cover the loan balance. In
the Potential for Losses these cases, the actual home value at the end of the loan is likely to be
more than what HUD assumes and therefore more likely to exceed the
loan balance at the end of the loan. According to HUD, because of this
conservative approach to estimating costs, the HECM program does not
rely on loans with property values that exceed the maximum claim amount
to operate on a break-even basis over the long-run.
Higher loan limits enacted under HERA and ARRA may make HUD’s
approach less conservative by reducing the proportion of loans for which
the property value exceeds the maximum claim amount. This scenario is
especially likely in locations that previously had relatively low local loan
limits (reflecting their lower home values) but are now subject to the
higher national limit. To illustrate, consider a 65-year-old HECM borrower
with a $400,000 home whose loan limit prior to HERA was $250,000 (see
fig. 12). In this scenario, the maximum claim amount would be the same as
the loan limit because the maximum claim amount is defined as the lesser
of the loan limit or the home value. However, if the loan limit for the same
borrower is increased to the HERA-authorized level of $417,000, the
maximum claim amount is the same as the home value ($400,000). As
figure 12 shows, when a borrower’s maximum claim amount is capped by
the loan limit, the maximum claim amount can be substantially lower than
the value of the home. All other things being equal, the potential for losses
is low in this scenario because the projected loan balance is likely to
remain less than the projected home value after the lender assigns the loan
to HUD. 30 In contrast, when the maximum claim amount is capped by the
home’s value, the difference between the projected loan balance and the
projected home value is smaller. The potential for losses is higher with
such a loan because the projected loan balance is more likely to exceed
the projected home value. As also shown in figure 12, when this effect is
combined with declining home prices, the potential for losses increases.
30
As previously noted, lenders may assign the loan to HUD when the loan balance reaches
98 percent of the maximum claim amount. At that point, HUD takes over servicing of the
loan and is responsible for covering any losses.

Figure 12: Illustration of How Increasing the HECM Loan Limit Could Increase
HUD’s Losses
Pre-HERA Post-HERA
Dollars (in thousands) Dollars (in thousands)

1,400 1,400
1,200 1,200
1,000 1,000
800 800
ce
lan
600 e 600 e ba
alu alu
ty v ty v an
per e pe
r
Lo
Pro lanc Pro $400,000
400 n ba 400
a max claim amount
Lo Loan
$250,000
assigned
200 max claim amount 200 to HUD
Loan
assigned
to HUD
0 0
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Age of loan in years Age of loan in years
Short-term decline in property

values at time of origination
No loss area (property value exceeds loan balance)

Loss area (loan balance exceeds property value)
Source: GAO.
Note: The trend line for property value assumes house price appreciation of 4 percent annually—this
is the rate HUD assumes in its pricing of the HECM loan product. The initial loan amount is based on
a $400,000 home, as well as an expected interest rate of 5.43 percent and a borrower age of 65,
which corresponds to a principal limit factor of 64.9 percent. The trend line for loan balance assumes
that the borrower immediately drew down 100 percent of the available loan funds.
Studies by HUD and others have noted that HECM loans for which the
home value exceeds the maximum claim amount have a positive impact on
the program’s financial performance but also have noted the potential
negative impact of raising the loan limit. When the HECM program started
in 1990, HUD developed a statistical model to estimate borrower payments
and insurance risk. HUD’s technical explanation of the model
acknowledges that future expected losses are smaller for HECMs with a

maximum claim amount capped by the loan limit, as compared with
HECMs with a maximum claim amount equal to the home value. 31
Similarly, actuarial reviews of the HECM program—conducted in 1995,
2000, and 2003—concluded that the negative net liability of the HECM
program resulted from homes valued at more than the HECM loan limit
cross-subsidizing those valued at less than the limit. 32 The 2003 actuarial
review also examined how the financial condition of the HECM program
would have been affected had a higher, national loan limit been in place
when existing HECMs were originated. The analysis found that the higher
loan limits would have reduced the expected net liability of the HECM
program from -$54.0 million to -$11.4 million. This finding is consistent
with a Congressional Budget Office (CBO) analysis of a 2007 legislative
proposal to increase the HECM loan limit to $417,000 nationwide. CBO
concluded that the increase would reduce HUD’s credit subsidy rate for
the 2008 cohort of loans from -1.9 percent to -1.35 percent. 33
The percentage of HECMs with maximum claim amounts capped by the

loan limit has declined in recent years (see fig. 13). Since the inception of
the program, this percentage has ranged from 24 percent to 47 percent.
However, this proportion has declined in recent years, dropping from 42
percent in fiscal year 2006 to 25 percent in fiscal year 2008. Furthermore,
HUD data show that this proportion dropped to 18 percent for the first 4
months of fiscal year 2009, likely due in part to the higher loan limit.
31
HUD, The FHA Home Equity Conversion Mortgage Insurance Demonstration: A Model
to Calculate Borrower Payments and Insurance Risk (Washington, D.C., October 1990).
32
HUD, Evaluation of the Home Equity Conversion Mortgage Insurance Demonstration:
Report to Congress. (Washington, D.C., 1995); Evaluation Report of FHA’s Home Equity
Conversion Mortgage Insurance Demonstration. Prepared by Abt. Assoc. (Washington,
D.C., Mar. 31, 2000); Refinancing Premium, National Loan Limit, and Long-Term Care
Premium Waiver For FHA’s HECM Program (Final Report). Prepared by Abt. Assoc.
(Washington, D.C., May 2003).
33
CBO, Cost Estimate: FHA Modernization Act of 2007 (Washington, D.C., Oct. 12, 2007).

Figure 13: Percentage of HECMs with Maximum Claim Amounts Limited by the Program Limit
Percentage
100
90 18
27 25 24 25
32 29 31 31 31
40 38 39 34 36
80 39 41 42
48 47
70
60
50
82
40 73 76 75
71 75
68 69 69 69
30 60 61 62 61 66 64 59
52 53 58
20
10
0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Fiscal year
Maximum claim amounts capped by HECM loan limit
Maximum claim amounts capped by value of property
Source: GAO analysis of HUD data.
Note: Fiscal year 2009 data include HECMs insured from October 1, 2008, through January 31, 2009.
HUD officials acknowledged that a reduction in the proportion of loans

with maximum claim amounts capped by the loan limit could have a
negative effect on the program’s financial performance. However, they
also indicated that their conservative approach to estimating program
costs mitigates the associated risks.

We provided a draft of this report to HUD for its review and comment. In
Agency Comments comments provided to us in an e-mail, HUD concurred with our report and
and Our Evaluation provided a technical comment, which we incorporated into the report.
We are sending copies of this report to interested congressional parties,

the Secretary of the Department of Housing and Urban Development, and
other interested parties. In addition, the report will be available at no
charge on our Web site at http://www.gao.gov. Contact points for our
Offices of Congressional Relations and Public Affairs may be found on the
last page of this report. If you or your staff has any questions about this
report, please contact me at (202) 512-8678 or sciremj@gao.gov. GAO
contact information and staff acknowledgments are listed in appendix IV.
Mathew J. Scirè
Director, Financial Markets
and Community Investment

Appendix I: Objectives, Scope, and
Methodology
Methodology
Our objectives were to examine (1) how the Housing and Economic
Recovery Act of 2008 (HERA) changes to the Home Equity Conversion
Mortgage (HECM) program and other factors have affected HECM lenders’
planned participation in the reverse mortgage market, (2) the extent to
which HERA’s changes to HECM origination fees and loan limits will
affect costs to borrowers and the loan amounts available to them, and (3)
Department of Housing and Urban Development’s (HUD) actions to
evaluate the financial performance of the HECM program, including the
potential impact of loan limit and house price changes.
To address these objectives, we reviewed laws, regulations and guidance

relevant to the HECM program, including provisions in HERA, the
American Recovery and Reinvestment Act of 2009 (ARRA), and HUD
handbooks and mortgagee letters. We also spoke with agency, industry,
and nonprofit officials, including those at HUD, Ginnie Mae, Fannie Mae,
the National Reverse Mortgage Lenders Association (NRMLA), the
Mortgage Bankers Association (MBA), and AARP.
To determine how HERA’s provisions have affected lenders’ planned

participation in the reverse mortgage market, we spoke with industry and
nonprofit officials—including those at Ginnie Mae, Fannie Mae, AARP,
NRMLA, and MBA—to understand how recent legislative and economic
changes were affecting the industry. To more specifically identify the
influence of legislation and economic factors on HECM lenders, we
conducted a Web-based survey of a random probability sample of the
2,779 lenders that originated HECMs on a retail basis in fiscal year 2008.
We used HUD records of HECM-certified lenders making at least one such
loan in fiscal year 2008, and supplemented HUD’s loan company officer
contact information with names and e-mail addresses of officers at those
lenders in our sample who also had memberships in NRMLA. For the
remaining sampled lenders for which we lacked contact information, we
made telephone calls to identify the most appropriate recipient for our
survey invitation.
We drew a stratified sample, allocating our selections across three groups

defined by the number of HECMs made in fiscal year 2008, sampling from
the groups with larger lenders at a higher rate than from the groups with
smaller lenders (see table 2). We sampled all 51 members of the stratum
with the largest lenders (300 or more loans). We sampled so few (30) and
received so few usable responses (8) from the stratum with the smallest
lenders (1 to 9 loans), that we considered this a nongeneralizable sample
and excluded it from our quantitative analysis. In addition, lenders in the
smallest lender stratum account for less than 5 percent of all loans, and

Methodology
thus would not influence overall estimates very much. Responses from the
smallest lenders stratum were used only as case study examples in our
analysis.
Table 2: Survey Population and Sample Dispositions
Fiscal
year 2008 Original Total Ineligible Usable Non- Response
a b
Stratum Population HECMs sample responses responses responses responses e rate
Very small (1-9
HECM loans in
2008)c 1,837 5,549 30 10 2 8 20 .80 33.3
Small (10 to 59 669 16,132 170 83 3 80 87 .96 48.8
loans in 2008)
Medium (60 to 222 27,247 113 66 13 53 47 .80 58.4
299 loans in
2008)
Large (300 or 51 63,087 51 41 2 39 10 .95 80.8
more loans in
2008)
Totalc 2,779 112,015 364 200 20 180 164 .91 56.9%c
Source: GAO.
a
e = estimated eligibility rate of nonresponding sample cases whose eligibility is unknown.
b
Response rate is unweighted, as defined by AAPOR RR3.
c
“Very small” stratum not included in overall response rate calculations.
To help develop our questionnaire, we consulted with an expert at

NRMLA. We pretested our draft questionnaire to officials at three HECM
lenders in our population and made revisions to it before finalization.
Legal and survey research specialists in GAO also reviewed the
questionnaire.
Before the survey, in early March 2009, NRMLA sent letters to those
lenders in our sample who were also members in that organization,
endorsing our survey and encouraging response. In March 2009, we sent e-
mails with links to our Web questionnaire and unique login information to
each member of our sample with valid e-mail addresses. For sampled
companies for which we were unable to obtain working e-mail addresses,
we mailed paper versions of the questionnaires. Nonresponding lenders
were sent additional e-mails or copies of questionnaires from March
through May. We also made telephone calls in April to nonrespondents
encouraging them to respond. Our survey closed in early May 2009. We

Methodology
received a total of 180 usable responses, for an overall response rate of 57

percent. 1 The “weighted” response rate for the survey, which takes into
account the relative numbers of lenders in the population that sampled
lenders in each of our three size strata had to represent, was 53 percent.
The most common reason for ineligibility among our sample firms was
closure, merger, or other discontinuation of business in the reverse
mortgage industry.
Because we followed a probability procedure based on random selections,

our sample is only one of a large number of samples that we might have
drawn. Since each sample could have provided different estimates, we
express our confidence in the precision of our particular sample’s results
as a 95 percent confidence interval (e.g., plus or minus 10 percentage
points). This is the interval that would contain the actual population value
for 95 percent of the samples we could have drawn. As a result, we are 95
percent confident that each of the confidence intervals in this report will
include the true values in the study population. Unless otherwise noted,
our estimates have margins of error of plus or minus 10 percentage points
or less at the 95 percent confidence interval.
In addition to sampling error, the practical difficulties of conducting any

survey may introduce other errors:
1. Nonresponse—bias from failing to get reports from lenders whose

answers would have differed significantly from those who did
participate.
2. Coverage—failure to include all eligible HECM lenders in the list from
which we sampled, or including ineligible firms.
3. Measurement—errors in response.
4. Data processing.
We took steps in developing the questionnaire, collecting the data, and

analyzing them to minimize such errors. For example, our pretesting and
expert reviews of the questionnaire resulted in question changes that
reduced the possibility of measurement error, and all data processing and
analysis programming was verified by independent analysts. In addition,
1
We used the response rate definition “RR3,” as defined by the American Association for
Public Opinion Research in “Standard Definitions: Final Dispositions of Case Codes and
Outcome Rates for Surveys,”
http://www.aapor.org/uploads/Standard_Definitions_07_08_Final.pdf, p. 35.

Methodology
we followed up on some unlikely answers by recontacting sampled

lenders or conducting followup research on them to edit erroneous
answers and declare some firms ineligible for our survey, thereby reducing
measurement and coverage error.
To assess the risk of nonresponse bias, we compared the response rates of

lenders across categories of two characteristics that might be related to
our key variables—the effect of HERA changes and other factors on the
likelihood of continuation of HECM lending in the future. The two
characteristics known for both respondents and nonrespondents were the
number of years the lender had been offering HECMs and the state in
which the lender’s home office is located, from which we could develop a
measure of size of loan activity in each state by summing the number of
loans made by lenders whose home offices were in a given state. We found
no statistically significant association between these two characteristics
and the likelihood of response. 2 Although this does not eliminate the
possibility of nonresponse bias, we found no evidence of bias based on our
analysis of this available data.
To determine the effect of the HERA provisions on HECM borrowers, we

examined changes in the up-front mortgage insurance premium,
origination fee, and loan funds available to borrowers. The up-front
mortgage insurance premium is 2 percent of the maximum claim amount.
HERA did not change this rate, but because of HERA’s change to the
HECM loan limit, some borrowers may be eligible for larger loans and
therefore have higher maximum claim amounts. Since the premium is
calculated based on the maximum claim amount, these borrowers will pay
a higher up-front mortgage insurance premium than they would have prior
to HERA. Before HERA, the origination fee was calculated as 2 percent of
the maximum claim amount with a minimum fee of $2,000. HERA changed
the calculation of the origination fee to 2 percent of the first $200,000 of
the maximum claim amount plus 1 percent of the maximum claim amount
over $200,000, with a maximum fee of $6,000. In implementing HERA,
HUD also increased the minimum origination fee by $500 to $2,500.
2
We grouped the sample lenders into two groups defined by when they were first approved
to offer HECM loans (before 2000 and 2000 or later), and found no association between the
year category and whether the lender responded. In addition, we compared sampled
lenders in the 5 states with the most HECM loans to the rest of the sampled lenders in the
remaining states and also found no association with whether the lender responded.

Methodology
We used two different approaches to assess the impact of the HERA

changes. First, we performed a mathematical analysis showing the
difference between the up-front costs before and after HERA. Specifically,
we derived equations for calculating pre-HERA and post-HERA up-front
costs for borrowers with maximum claim amounts in different ranges ($0
to $100,000; $100,000 to $125,000; $125,000 to $200,000; $200,000 to
$400,000; and $400,000 to $625,500). 3 For each range, we subtracted the
pre-HERA equation from the post-HERA equation to derive an equation for
calculating the change in up-front costs due to the HERA provisions. We
then used these equations to calculate the potential change in up-front
costs in dollars terms. We did this analysis separately for cases in which
the maximum claim amount would increase under HERA and cases in
which the maximum claim amount would remain the same. Appendix III
shows the details of this analysis.
Second, we applied the HERA changes to HUD loan-level data for HECMs
that borrowers obtained in calendar year 2007. We compared the results to
the actual up-front costs and loan funds available for these borrowers. To
perform this analysis, we obtained data from HUD’s Single-family Data
Warehouse. We assessed the reliability of these data by (1) reviewing
existing information about the data and the system that produced them,
(2) interviewing HUD officials knowledgeable about the data, and (3)
performing electronic testing of required data elements. We determined
that the data we used were sufficiently reliable for the purposes of this
report. As shown in table 3, the universe of 2007 HECMs used in our
analysis included 101,480 loans. 4 We applied the $417,000 national loan
limit and HERA’s changes to the origination fee calculation to the 2007
HECMs. For each borrower, we calculated the new maximum claim
amount, origination fee, up-front mortgages insurance premium, and loan
funds available under the HERA rules and compared our results to the
actual 2007 values. We summarized our results by calculating the average
changes in these amounts.
3
We used $625,500 as a maximum because ARRA resulted in an increase in the HECM loan
limit to that amount.
4
The 101,480 loans did not include 6,664 HECMs obtained in 2007 to refinance an existing
HECM. We analyzed refinanced HECMs separately because the calculation for the
mortgage insurance premium is different for these loans than for new HECMs. Our results
for the analysis of refinanced HECMs were similar to those for new HECMs.

Methodology
Table 3: Universe of 2007 HECMs by Home Value and Maximum Claim Amount
Category
Maximum claim amount Maximum claim amount

Home value equals loan limit equals home value Total
Under $125,000 0 17,303 17,303
(17%)
$125,000 to $200,000 6 27,548 27,554)
(27%)
$200,001 to $399,999 14,535 27,474 42,009
(41%)
$400,000 or more 14,433 181 14,614
(14%)
Total 28,974 72,506 101,480
(29%) (71%) (100%)
Source: GAO analysis of HUD data.
Note: Percentages do not add to 100 because of rounding.
To illustrate the potential effect of modest margin rates increases

stemming from HERA’s change to the origination fee calculation, we
applied a 0.25 percentage point increase to the margin rate for the 2007
HECMs adjusted to reflect the HERA provisions. 5 We determined the
resulting changes in the loan funds available to borrowers using HUD’s
table of principal limit factors. To provide perspective on the HERA-
related margin rate changes, we compared margin rates from a 3 month
period 1 year prior to the implementation of HERA (November 2007
through January 2008) to the margin rates from the 3 month period after
the implementation of HERA (November 2008 through January 2009).
To examine HUD’s actions to evaluate the financial performance of the

HECM program, we reviewed HUD’s budget estimates for the HECM
program for fiscal years 2005 through 2010. We also compiled and
analyzed financial performance information about the HECM program,
including the liability for loan guarantee (LLG) and credit subsidy
5
As discussed in the body of this report, our survey of HECM lenders indicated that some
lenders have raised their margin rates modestly to compensate for HERA’s limitations on
the origination fee. We did not receive a sufficient number of responses to reliably estimate
the average increase in margin rate for the population. Of the lenders that responded to the
survey question, we estimate that 48 percent increased margin rates for HECMs offered by
their institution by at least 0.25 percentage points. This estimate has a margin of error of
within plus or minus 13 percentage points.

Methodology
estimates. For example, we examined the Federal Housing

Administration’s (FHA) Annual Management Reports (2005, 2006, 2007,
and 2008), which include FHA’s annual financial statements; HUD Office
of the Inspector General (OIG) audits of FHA’s financial statements (2005,
2006, 2007, and 2008); actuarial reviews of the HECM program (1995, 2000,
and 2003); and Congressional Budget Office cost estimates relevant to the
HECM program. We also reviewed other analyses HUD has conducted of
program costs, such as the sensitivity of estimated cash flows to
alternative economic assumptions. We interviewed FHA officials about
their budget estimates and program analyses. Additionally, we reviewed
information about HUD’s HECM cash flow model, including a technical
explanation of the model published in 1990 and recent changes to the
model. We also reviewed historical house price appreciation rates from
the Federal Housing Finance Agency and projected house price
appreciation rates from IHS Global Insight. To examine the percentage of
HECMs with maximum claim amounts capped by the loan limit, we
analyzed loan-level data on HECMs from HUD’s Single-family Data
Warehouse. As noted earlier, we determined that the data we used were
sufficiently reliable for this analysis. In addition, we reviewed federal
agency standards for managing credit programs, such as those contained
in the Federal Credit Reform Act (FCRA), related Office of Management
and Budget requirements and instructions, and Federal Accounting
Standards Advisory Board guidance. Finally, we interviewed HUD OIG
officials, industry participants, and mortgage market analysts.
We conducted this performance audit from September 2008 through July

2009, in accordance with generally accepted government auditing
standards. Those standards require that we plan and perform the audit to
obtain sufficient, appropriate evidence to provide a reasonable basis for
our findings and conclusions based on our audit objectives. We believe
that the evidence obtained provides a reasonable basis for our findings
and conclusions based on our audit objectives.

Appendix II: Impact of Loan Limit Increase in
Appendix II: Impact of Loan Limit Increase in the American Recovery and Reinvestment Act
of 2009 on HECM Lenders
the American Recovery and Reinvestment

Act of 2009 on HECM Lenders
The American Recovery and Reinvestment Act (ARRA) raised the national
loan limit for Home Equity Conversion Mortgages (HECM) to $625,500
through December 31, 2009. In our survey of HECM lenders, we asked
lenders about the influence the increased loan limit would have on their
likelihood to offer HECMs and non-HECM reverse mortgages (see fig. 14).
Additionally, we asked how they expected consumer demand for HECMs
to increase as a result of the ARRA loan increase (see fig. 15). See figures
14 and 15 for survey questions and estimates based on our survey results.
Figure 14: Influence of ARRA’s Increase to Loan Limits on Lenders’ Plans to Offer Reverse Mortgages
Moderate or great Little or no Great to moderate Don’t Not

upward influence influence downward influence know applicable
Plans to offer HECMs 65% 34 % 0% 1% N/A %
Plans to offer non-HECM reverse

mortgages 29 54 11 6 N/A
Figure 15: Influence of ARRA’s Increase to Loan Limits on Consumer Demand for HECMs
Increase somewhat Decrease somewhat Don’t Not

or greatly No effect or greatly know applicable
Lenders’ expectation of ARRA’s

effect on consumer demand for 82 % 13 % 0% 5% N/A %
HECMs

Appendix III: Effect of the Housing and
Economic Recovery Act of 2008 on Up-front
Costs for HECM Borrowers

Home Equity Conversion Mortgage (HECM) borrowers may experience
changes in up-front costs due to the Housing and Economic Recovery Act
of 2008’s (HERA) change to the calculation of the origination fee, the loan
limit, or both. Generally, borrowers with house values greater than the
prior HECM loan limit will be able to borrow more under HERA’s higher
loan limit, while borrowers with a wide range of house values may be
affected by the changes in origination fees. 1
There are two up-front costs. The first—the up-front mortgage insurance
premium—is 2 percent of the maximum claim amount. The second—the
origination fee—was calculated before HERA as 2 percent of the
maximum claim amount with a minimum fee of $2,000. HERA changed the
calculation of the origination fee to 2 percent of the first $200,000 of the
maximum claim amount plus 1 percent of the maximum claim amount
over $200,000, with a maximum fee of $6,000. In implementing HERA,
HUD also increased the minimum origination fee by $500 to $2,500.
To determine how borrowers would be affected by these changes, we

developed mathematical equations for calculating the up-front costs under
both the HERA and pre-HERA rules. We subtracted the equation for the
pre-HERA rules from the equation for the HERA rules to derive an
equation for the change in up-front costs resulting from HERA. A positive
value indicates that a borrower would pay more under HERA, and a
negative value indicates that a borrower would pay less. Figures 16 and 17
illustrate how these changes affect different categories of borrowers.
Figure 16 shows the results for borrowers who have home values lower
than the previous loan limit. The maximum claim amount is not affected
by HERA’s change in loan limit. Therefore, for these borrowers, changes in
up-front costs derive only from changes in the origination fee.
1
HERA did not change the HECM loan limits in parts of Hawaii that had loan limits higher
than $417,000 prior to HERA. As a result, borrowers in those areas whose house values are
greater than the pre-HERA loan limits will see no change in their up-front mortgage
insurance premiums due to HERA.

Figure 16: Changes to Up-front Costs for Borrowers Not Affected by HERA’s
Change in Loan Limit
Maximum claim Equation for change in up-front

amount (MC) costs due to HERA (dollars) Range of change to up-front costs
$0 to $100,000 500 $500 more
$100,000 to $125,000 2,500 - (0.02 X MC) from $0 to 500 more
$125,000 to $200,000 0 $0
$200,000 to $400,000 2,000 - (0.01 X MC) from $0 to $2,000 less
$400,000 to $625,500 6,000 - (0.02 X MC) from $2,000 to $6,510 less
Borrow pays more
Borrower pays the same or less
Source: GAO.
Figure 17 shows the results of the calculation for borrowers who were
affected by HERA’s increase in loan limit. These borrowers would pay up-
front mortgage insurance premiums and origination fees based on a higher
maximum claim amount. However, depending on the maximum claim
amount, the origination fee may have decreased rather than increased. The
net change in up-front costs for this grouping is therefore indeterminable
without knowing the old and new maximum claim amounts.
Figure 17: Changes to Up-front Costs for Borrowers Affected by HERA’s Change in Loan Limit
Note: New maximum claim amounts and old maximum claim amounts less than $125,000 are not
valid because no local loan limits were less than $125,000.

Appendix IV: GAO Contact and Staff
Appendix IV: GAO Contact and Staff
Acknowledgments
Acknowledgments
Mathew J. Scirè, at (202) 512-8678 or sciremj@gao.gov

GAO Contact
In addition to the individual named above, Steve Westley, Assistant
Staff Director; Anne Akin, Kathleen Boggs, Joanna Chan, Rudy Chatlos, Karen
Acknowledgments Jarzynka, John McGrail, Marc Molino, Mark Ramage, Carl Ramirez,
Barbara Roesmann, and Jennifer Schwartz made key contributions to this
report.
(250421)
GAO’s Mission The Government Accountability Office, the audit, evaluation, and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO’s
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.
The fastest and easiest way to obtain copies of GAO documents at no cost
Obtaining Copies of is through GAO’s Web site (www.gao.gov). Each weekday afternoon, GAO
GAO Reports and posts on its Web site newly released reports, testimony, and
correspondence. To have GAO e-mail you a list of newly posted products,
Testimony go to www.gao.gov and select “E-mail Updates.”
Order by Phone The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site,
http://www.gao.gov/ordering.htm.
Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537.
Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional information.
Contact:
To Report Fraud,
Waste, and Abuse in Web site: www.gao.gov/fraudnet/fraudnet.htm
E-mail: fraudnet@gao.gov
Federal Programs Automated answering system: (800) 424-5454 or (202) 512-7470
Ralph Dawn, Managing Director, dawnr@gao.gov, (202) 512-4400

Congressional U.S. Government Accountability Office, 441 G Street NW, Room 7125
Relations Washington, DC 20548
Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800

Public Affairs U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, DC 20548
Please Print on Recycled Paper

GAO Report On Reverse Mortgages July 2009

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

GAO Report On Reverse Mortgages July 2009

Uploaded by

Copyright:

Available Formats

United States Government Accountability Office

GAO Report to Congressional Committees

Policy Changes Have

Why GAO Did This Study What GAO Found

View GAO-09-836 or key components.

United States Government Accountability Office

Appendix I Objectives, Scope, and Methodology 38

Appendix II Impact of Loan Limit Increase in the American

Appendix III Effect of the Housing and Economic Recovery Act of

Appendix IV GAO Contact and Staff Acknowledgments 48

Page i GAO-09-836 Reverse Mortgages

Page ii GAO-09-836 Reverse Mortgages

ARRA American Recovery and Reinvestment Act of 2009

Page iii GAO-09-836 Reverse Mortgages

July 30, 2009

The Honorable Christopher Dodd

The Honorable Barney Frank

Page 1 GAO-09-836 Reverse Mortgages

In light of these changes, HERA contains a mandate for GAO to evaluate

To address these objectives, we reviewed laws, regulations, and guidance

Page 2 GAO-09-836 Reverse Mortgages

We conducted this performance audit from September 2008 through July

A reverse mortgage is a loan against the borrower’s home that the

Page 3 GAO-09-836 Reverse Mortgages

Figure 1: Comparison of 30-Year Forward and Reverse Mortgages

Dollars (in thousands) Traditional 30-year forward mortgage

Dollars (in thousands) Reverse mortgage, over 30 years

Loan balance (debt)

Page 4 GAO-09-836 Reverse Mortgages

Additionally, the potential liability of loans insured by FHA has doubled in

Page 5 GAO-09-836 Reverse Mortgages

Page 6 GAO-09-836 Reverse Mortgages

Lenders can participate in the HECM market through wholesale or retail

There is a secondary market for HECMs, as most lenders prefer not to

The amount of loan funds available to the borrower is determined by

Page 7 GAO-09-836 Reverse Mortgages

• Third, the funds available to the borrower are further reduced by a

Page 8 GAO-09-836 Reverse Mortgages

Borrowers incur various costs when obtaining a HECM. HUD allows

• Origination fee: Prior to HERA, HECM borrowers were charged an

• Mortgage insurance premium: Borrowers are charged an up-front

• Interest: Borrowers are charged interest, which generally includes a base

Page 9 GAO-09-836 Reverse Mortgages

• HECM counseling fee: The HECM program requires prospective

• Loan servicing fee: Borrowers pay a monthly servicing fee of up to $35.

FCRA requires federal agencies that provide loan guarantees to estimate

Page 10 GAO-09-836 Reverse Mortgages

Generally, agencies are required to produce annual updates of their

In combination, HERA’s changes to the HECM loan limit and origination

Page 11 GAO-09-836 Reverse Mortgages

Page 12 GAO-09-836 Reverse Mortgages

HERA's changes to the

HERA's changes to both HECM

Current economic conditions 52 28 19 0 0

Reduced opportunities in the

HERA's prohibition of participation

Current secondary market options 14 50 32 0 4

Current house price trends 10 50 38 1 1

HERA's restrictions on selling

Source: GAO analysis of survey of HECM lenders.

Page 13 GAO-09-836 Reverse Mortgages

We estimate that only 2 percent of HECM lenders do not plan to continue

As part of our survey, we asked lenders how various economic and